高铁接触网旋转双耳销钉状态检测方法研究
发布时间：2018-08-01 17:24
【摘要】：针对高速铁路接触网支撑装置旋转双耳部件销钉的松脱与脱落问题,提出一种基于图像不变性目标定位及灰度分布规律特征的销钉不良状态检测方法。通过分析现场接触网支撑及悬挂装置图像,利用SIFT(Scale Invariant Feature transform)算法和改进的RANSAC(Random Sample Consensus)算法实现双耳部件的定位;采用Hough变换实现目标图像中双耳套筒倾角的提取,并将其旋转至水平方向,进而实现旋转双耳部分的分割;累加目标图像的竖直方向像素灰度值,确定销钉受力部分和两端非受力部分长度;归纳销钉正常工作及故障时这些长度间相关比值的范围,从而判断销钉的工作状态。实验表明,该方法能够较准确地实现销钉不良状态的检测。
[Abstract]:In order to solve the problem of loosening and shedding of pin of rotary binaural parts in catenary support device of high-speed railway, a new method for detecting the bad state of pin based on image invariance target location and gray distribution rule is proposed. By analyzing the images of field catenary support and suspension device, the location of binaural parts is realized by using SIFT (Scale Invariant Feature transform) algorithm and improved RANSAC (Random Sample Consensus) algorithm, and the obliquity of binaural sleeve in target image is extracted by Hough transform. It rotates to the horizontal direction, then realizes the segmentation of the rotating binaural part, accumulates the pixel gray value in the vertical direction of the target image, and determines the length of the pin bearing part and the unloaded part at both ends. The range of the correlation ratio between these lengths when the pin is in normal operation and fault is summarized to judge the working state of the pin. Experiments show that the method can accurately detect the bad state of pin.
【作者单位】： 西南交通大学电气工程学院;
【基金】：国家自然科学基金(U1434203,51377136,51407147)
【分类号】：TP391.41;U226.8
本文编号：2158298
[Abstract]:In order to solve the problem of loosening and shedding of pin of rotary binaural parts in catenary support device of high-speed railway, a new method for detecting the bad state of pin based on image invariance target location and gray distribution rule is proposed. By analyzing the images of field catenary support and suspension device, the location of binaural parts is realized by using SIFT (Scale Invariant Feature transform) algorithm and improved RANSAC (Random Sample Consensus) algorithm, and the obliquity of binaural sleeve in target image is extracted by Hough transform. It rotates to the horizontal direction, then realizes the segmentation of the rotating binaural part, accumulates the pixel gray value in the vertical direction of the target image, and determines the length of the pin bearing part and the unloaded part at both ends. The range of the correlation ratio between these lengths when the pin is in normal operation and fault is summarized to judge the working state of the pin. Experiments show that the method can accurately detect the bad state of pin.
【作者单位】： 西南交通大学电气工程学院;
【基金】：国家自然科学基金(U1434203,51377136,51407147)
【分类号】：TP391.41;U226.8
【相似文献】
相关期刊论文 前5条
1 谢拴勤;郭振安;;基于热棒工作状态检测问题的分析[J];计算机测量与控制;2008年07期
2 ;2002年12月份三大干线轨道状态检测结果简报(0212号)[J];铁道建筑;2003年03期
3 ;2002年11月份三大干线轨道状态检测结果简报(0211号)[J];铁道建筑;2003年02期
4 余先元;康太平;郭成宝;;基于Nios处理器的磁浮车运动状态检测与控制[J];四川理工学院学报(自然科学版);2006年02期
5 ;[J];;年期
相关硕士学位论文 前3条
1 惠美玲;基于四元数小波变换的结构状态检测研究[D];大连理工大学;2016年
2 田安刚;列车司控运停状态检测分析系统研究[D];青岛科技大学;2012年
3 杨红梅;基于图像处理的接触网支持及悬挂装置不良状态检测[D];西南交通大学;2013年
，本文编号：2158298
本文链接：https://www.wllwen.com/kejilunwen/ruanjiangongchenglunwen/2158298.html
